qtextcodec.cpp

doxygen(一个自动从源代码生成文档的工具)的源代码

                        cursor[3]>='0' && cursor[3]<='9') {

                        unicode = strtol(cursor+2,&cursor,16);

                    } else if (*cursor==esc) {

                        byte = getByte(cursor);

                        if ( *cursor == esc ) {
                            if ( !to_unicode_multibyte ) {
                                to_unicode_multibyte =
                                    new QMultiByteUnicodeTable[256];
                                for (int i=0; i<256; i++) {
                                    to_unicode_multibyte[i].unicode =
                                        to_unicode[i];
                                    to_unicode_multibyte[i].multibyte = 0;
                                }
                                delete [] to_unicode;
                                to_unicode = 0;
                            }
                            QMultiByteUnicodeTable* mbut =
                                to_unicode_multibyte+byte;
                            mb[nmb++] = byte;
                            while ( nmb < maxmb && *cursor == esc ) {
                                // Always at least once

                                mbut->unicode = CHAINED;
                                byte = getByte(cursor);
                                mb[nmb++] = byte;
                                if (!mbut->multibyte) {
                                    mbut->multibyte =
                                        new QMultiByteUnicodeTable[256];
                                }
                                mbut = mbut->multibyte+byte;
                                mb_unicode = & mbut->unicode;
                            }

                            if ( nmb > max_bytes_per_char )
                                max_bytes_per_char = nmb;
                        }
                    } else {
                        cursor++;
                    }
                }

                if (unicode >= 0 && unicode <= 0xffff)
                {
                    QChar ch((ushort)unicode);
                    if (!from_unicode_page[ch.row()]) {
                        from_unicode_page[ch.row()] = new char[256];
                        for (int i=0; i<256; i++)
                            from_unicode_page[ch.row()][i]=0;
                    }
                    if ( mb_unicode ) {
                        from_unicode_page[ch.row()][ch.cell()] = 0;
                        if (!from_unicode_page_multibyte) {
                            from_unicode_page_multibyte = new char**[256];
                            for (int i=0; i<256; i++)
                                from_unicode_page_multibyte[i]=0;
                        }
                        if (!from_unicode_page_multibyte[ch.row()]) {
                            from_unicode_page_multibyte[ch.row()] = new char*[256];
                            for (int i=0; i<256; i++)
                                from_unicode_page_multibyte[ch.row()][i] = 0;
                        }
                        mb[nmb++] = 0;
                        from_unicode_page_multibyte[ch.row()][ch.cell()]
                            = qstrdup(mb);
                        *mb_unicode = unicode;
                    } else {
                        from_unicode_page[ch.row()][ch.cell()] = (char)byte;
                        if ( to_unicode )
                            to_unicode[byte] = unicode;
                        else
                            to_unicode_multibyte[byte].unicode = unicode;
                    }
                } else {
                }
            }
        }
        n = n.stripWhiteSpace();

        unkn = '?'; // ##### Might be a bad choice.
    }

    ~QTextCodecFromIOD()
    {
        if ( from_unicode_page ) {
            for (int i=0; i<256; i++)
                if (from_unicode_page[i])
                    delete [] from_unicode_page[i];
        }
        if ( from_unicode_page_multibyte ) {
            for (int i=0; i<256; i++)
                if (from_unicode_page_multibyte[i])
                    for (int j=0; j<256; j++)
                        if (from_unicode_page_multibyte[i][j])
                            delete [] from_unicode_page_multibyte[i][j];
        }
        if ( to_unicode )
            delete [] to_unicode;
        if ( to_unicode_multibyte )
            delete [] to_unicode_multibyte;
    }

    bool ok() const
    {
        return !!from_unicode_page;
    }

    QTextDecoder* makeDecoder() const
    {
        if ( stateless() )
            return QTextCodec::makeDecoder();
        else
            return new QTextCodecFromIODDecoder(this);
    }

    const char* name() const
    {
        return n;
    }

    int mibEnum() const
    {
        return 0; // #### Unknown.
    }

    int heuristicContentMatch(const char*, int) const
    {
        return 0;
    }

    int heuristicNameMatch(const char* hint) const
    {
        int bestr = QTextCodec::heuristicNameMatch(hint);
        QStrListIterator it(aliases);
        char* a;
        while ((a=it.current())) {
            ++it;
            int r = simpleHeuristicNameMatch(a,hint);
            if (r > bestr)
                bestr = r;
        }
        return bestr;
    }

    QString toUnicode(const char* chars, int len) const
    {
        const uchar* uchars = (const uchar*)chars;
        QString result;
        QMultiByteUnicodeTable* multibyte=to_unicode_multibyte;
        if ( multibyte ) {
            while (len--) {
                QMultiByteUnicodeTable& mb = multibyte[*uchars];
                if ( mb.multibyte ) {
                    // Chained multi-byte
                    multibyte = mb.multibyte;
                } else {
                    result += QChar(mb.unicode);
                    multibyte=to_unicode_multibyte;
                }
                uchars++;
            }
        } else {
            while (len--)
                result += QChar(to_unicode[*uchars++]);
        }
        return result;
    }

    QCString fromUnicode(const QString& uc, int& lenInOut) const
    {
        if (lenInOut > (int)uc.length())
            lenInOut = uc.length();
        int rlen = lenInOut*max_bytes_per_char;
        QCString rstr(rlen);
        char* cursor = rstr.data();
        char* s=0;
        int l = lenInOut;
        int lout = 0;
        for (int i=0; i<l; i++) {
            QChar ch = uc[i];
            if ( ch == QChar::null ) {
                // special
                *cursor++ = 0;
            } else if ( from_unicode_page[ch.row()] &&
                from_unicode_page[ch.row()][ch.cell()] )
            {
                *cursor++ = from_unicode_page[ch.row()][ch.cell()];
                lout++;
            } else if ( from_unicode_page_multibyte &&
                      from_unicode_page_multibyte[ch.row()] &&
                      (s=from_unicode_page_multibyte[ch.row()][ch.cell()]) )
            {
                while (*s) {
                    *cursor++ = *s++;
                    lout++;
                }
            } else {
                *cursor++ = unkn;
                lout++;
            }
        }
        *cursor = 0;
        lenInOut = lout;
        return rstr;
    }
};

QTextCodecFromIODDecoder::QTextCodecFromIODDecoder(const QTextCodecFromIOD* c) :
    codec(c)
{
    mb = codec->to_unicode_multibyte;
}

QString QTextCodecFromIODDecoder::toUnicode(const char* chars, int len)
{
    const uchar* uchars = (const uchar*)chars;
    QString result;
    while (len--) {
        QMultiByteUnicodeTable& t = mb[*uchars];
        if ( t.multibyte ) {
            // Chained multi-byte
            mb = t.multibyte;
        } else {
            if ( t.unicode )
                result += QChar(t.unicode);
            mb=codec->to_unicode_multibyte;
        }
        uchars++;
    }
    return result;
}

/*!
  Reads a POSIX2 charmap definition from \a iod.
  The parser recognizes the following lines:
<pre>
   &lt;code_set_name&gt; <i>name</i>
   &lt;escape_char&gt; <i>character</i>
   % alias <i>alias</i>
   CHARMAP
   &lt;<i>token</i>&gt; /x<i>hexbyte</i> &lt;U<i>unicode</i>&gt; ...
   &lt;<i>token</i>&gt; /d<i>decbyte</i> &lt;U<i>unicode</i>&gt; ...
   &lt;<i>token</i>&gt; /<i>octbyte</i> &lt;U<i>unicode</i>&gt; ...
   &lt;<i>token</i>&gt; /<i>any</i>/<i>any</i>... &lt;U<i>unicode</i>&gt; ...
   END CHARMAP
</pre>

  The resulting QTextCodec is returned (and also added to the
  global list of codecs).  The name() of the result is taken
  from the code_set_name.

  Note that a codec constructed in this way uses much more memory
  and is slower than a hand-written QTextCodec subclass, since
  tables in code are in memory shared by all applications simultaneously
  using Qt.

  \sa loadCharmapFile()
*/
QTextCodec* QTextCodec::loadCharmap(QIODevice* iod)
{
    QTextCodecFromIOD* r = new QTextCodecFromIOD(iod);
    if ( !r->ok() ) {
        delete r;
        r = 0;
    }
    return r;
}

/*!
  A convenience function for loadCharmap().
*/
QTextCodec* QTextCodec::loadCharmapFile(QString filename)
{
    QFile f(filename);
    if (f.open(IO_ReadOnly)) {
        QTextCodecFromIOD* r = new QTextCodecFromIOD(&f);
        if ( !r->ok() )
            delete r;
        else
            return r;
    }
    return 0;
}
#endif //QT_NO_CODECS


/*!
  Returns a string representing the current language.
*/

const char* QTextCodec::locale()
{
    static QCString lang;
    if ( lang.isEmpty() ) {
        lang = getenv( "LANG" ); //########Windows??
        if ( lang.isEmpty() )
            lang = "C";
    }
    return lang;
}



#ifndef QT_NO_CODECS

class QSimpleTextCodec: public QTextCodec
{
public:
    QSimpleTextCodec( int );
    ~QSimpleTextCodec();

    QString toUnicode(const char* chars, int len) const;
    QCString fromUnicode(const QString& uc, int& lenInOut ) const;

    const char* name() const;
    int mibEnum() const;

    int heuristicContentMatch(const char* chars, int len) const;

    int heuristicNameMatch(const char* hint) const;

private:
    int forwardIndex;
};


#define LAST_MIB 2259

static struct {
    const char * cs;
    int mib;
    Q_UINT16 values[128];
} unicodevalues[] = {
    // from RFC 1489, ftp://ftp.isi.edu/in-notes/rfc1489.txt
    { "KOI8-R", 2084,
      { 0x2500, 0x2502, 0x250C, 0x2510, 0x2514, 0x2518, 0x251C, 0x2524,